Bug 191256 - Fix build with disabled DFG/FTL
Summary: Fix build with disabled DFG/FTL
Status: RESOLVED FIXED
Alias: None
Product: WebKit
Classification: Unclassified
Component: New Bugs (show other bugs)
Version: WebKit Nightly Build
Hardware: Unspecified Unspecified
: P2 Normal
Assignee: Nobody
URL:
Keywords:
Depends on:
Blocks: 191548
  Show dependency treegraph
 
Reported: 2018-11-05 03:32 PST by Dominik Inführ
Modified: 2018-11-17 02:34 PST (History)
10 users (show)

See Also:


Attachments
Patch (6.29 KB, patch)
2018-11-05 03:32 PST, Dominik Inführ
no flags Details | Formatted Diff | Diff
Patch (6.31 KB, patch)
2018-11-05 03:37 PST, Dominik Inführ
no flags Details | Formatted Diff | Diff
Patch (6.31 KB, patch)
2018-11-05 09:29 PST, Dominik Inführ
no flags Details | Formatted Diff | Diff

Note You need to log in before you can comment on or make changes to this bug.
Description Dominik Inführ 2018-11-05 03:32:24 PST
Fix build with disabled DFG/FTL[D[D[D[D[D[D[D[D[D[D[D[D[D[D[D[D[D[D[D[D[D[D[D[C
Comment 1 Dominik Inführ 2018-11-05 03:32:49 PST
Created attachment 353845 [details]
Patch
Comment 2 Dominik Inführ 2018-11-05 03:37:15 PST
Created attachment 353846 [details]
Patch
Comment 3 Guillaume Emont 2018-11-05 03:49:46 PST
Comment on attachment 353846 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=353846&action=review

Did not try it yet, but the changes seem to make sense.
One suggestion for improvement though: the implementation of GetByIdStatus::computeForStubInfoWithoutExitSiteFeedback() in bytecode/GetByIdStatus.cpp does not seem to be guarded by ENABLE(DFG_JIT) even though its definition in the .h is, so maybe we want to guard it too?

> Source/JavaScriptCore/bytecode/PutByIdStatus.cpp:-55
> -#if ENABLE(DFG_JIT)
> -ExitFlag PutByIdStatus::hasExitSite(CodeBlock* profiledBlock, unsigned bytecodeIndex)
> -{
> -    return hasBadCacheExitSite(profiledBlock, bytecodeIndex);
> -}
> -#endif

Any idea why this function existed in the first place?
Comment 4 Dominik Inführ 2018-11-05 05:51:18 PST
After looking into the history, it seems that the function used to do more stuff.
Comment 5 Guillaume Emont 2018-11-05 09:28:51 PST
(In reply to Guillaume Emont from comment #3)
> One suggestion for improvement though: the implementation of
> GetByIdStatus::computeForStubInfoWithoutExitSiteFeedback() in
> bytecode/GetByIdStatus.cpp does not seem to be guarded by ENABLE(DFG_JIT)
> even though its definition in the .h is, so maybe we want to guard it too?

Disregard that, I got mixed up between GetByIdStatus::computeForStubInfoWithoutExitSiteFeedback() and InByIdStatus::computeForStubInfoWithoutExitSiteFeedback().
Comment 6 Dominik Inführ 2018-11-05 09:29:16 PST
Created attachment 353873 [details]
Patch
Comment 7 Dominik Inführ 2018-11-08 01:45:36 PST
ping review.
Comment 8 Yusuke Suzuki 2018-11-17 02:07:05 PST
Comment on attachment 353873 [details]
Patch

r=me
Comment 9 WebKit Commit Bot 2018-11-17 02:33:24 PST
Comment on attachment 353873 [details]
Patch

Clearing flags on attachment: 353873

Committed r238346: <https://trac.webkit.org/changeset/238346>
Comment 10 WebKit Commit Bot 2018-11-17 02:33:25 PST
All reviewed patches have been landed.  Closing bug.